Abstract
The performance of image restoration depends on the choice of image priors and regularization parameters. Recent studies reveal that heavy-tailed sparse gradient distributions of natural images can be modeled by hyper-Laplacian with the norm p∈ [0.5, 1]. However, it is still very challenging to determine the values of the norm and the regularization parameters. In this paper, we proposed a maximum a posterior (MAP) method for adaptive regularization parameters and norm selection. During the procedure of image restoration, by modeling the estimated residual and gradient distribution with Gaussian and hyper-Laplacian, respectively, we suggest a MAP formulation for the joint estimation of the latent image, regularization parameter, and norm. Then, we propose an alternating optimization method to iteratively solving the MAP problem. Experimental results show that, the proposed method obtained satisfactory restoration results for various degraded images with different noise level, and outperformed the other methods.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Rudin, L.I., Osher, S., Fatemi, E.: Nonlinear Total Variation Based Noise Removal Algorithms. J. Physica D 60(1-4), 259–268 (1992)
Fergus, R., Singh, B., Hertzmann, A., Roweis, S., Freeman, W.T.: Removing Camera Shake from a Single Photograph. Proc. ACM 25(3), 787–794 (2006)
Levin, A., Weiss, Y., Durand, F., Freeman, W.T.: Efficient Marginal Likelihood Optimization in Blind Deconvolution. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 2657–2664. IEEE (2011)
Levin, A., Fergus, R., Durand, F., Freeman, W.T.: Image and Depth from a Conventional Camera with a Coded Aperture. ACM 26(3), 70 (2007)
Krishnan, D., Fergus, R.: Fast Image Deconvolution Using Hyper-Laplacian Priors. In: NIPS, vol. 22, pp. 1–9 (2009)
Cho, T.S., Joshi, N., Zitnick, C.L., et al.: A Content-Aware Image Prior. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 169–176 (2010)
Cho, T.S., Zitnick, C.L., Joshi, N., et al.: Image Restoration by Matching Gradient Distributions. IEEE Transactions on Pattern Analysis and Machine Intelligence 34(4), 683–694 (2012)
Donoho, D.L.: De-Noising by Soft-Thresholding. IEEE Transactions on Information Theory 41(3), 613–627 (1995)
Stefano, A., White, P.R., Collis, W.B.: Training Methods for Image Noise Level Estimation on Wavelet Components. J. EURASIP on Advances in Signal Processing 16, 2400–2407 (2004)
Galatsanos, N.P., Katsaggelos, A.K.: Methods for Choosing the Regularization Parameter and Estimating the Noise Variance in Image Restoration and their Relation. IEEE Transactions on Image Processing 1(3), 322–336 (1992)
Hansen, P.C.: Analysis of Discrete Ill-posed Problems by Means of the L-curve. SIAM Review 34(4), 561–580 (1992)
Wen, Y.W., Chan, R.H.: Parameter Selection for Total-Variation-Based Image Restoration Using Discrepancy Principle. IEEE Transactions on Image Processing 21(4), 1770–1781 (2012)
Geman, D., Yang, C.: Nonlinear Image Recovery with Half-quadratic Regularization. IEEE Transactions on Image Processing 4(7), 932–946 (1995)
Wang, Y., Yang, et al.: A New Alternating Minimization Algorithm for Total Variation Image Reconstruction. SIAM Journal on Imaging Sciences 1(3), 248–272 (2008)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Lin, X., Zhang, H., Deng, H., Zuo, W. (2013). Adaptive Regularization Parameters and Norm Selection for Sparse Gradient Based Image Restoration. In: Sun, C., Fang, F., Zhou, ZH., Yang, W., Liu, ZY. (eds) Intelligence Science and Big Data Engineering. IScIDE 2013. Lecture Notes in Computer Science, vol 8261. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-42057-3_99
Download citation
DOI: https://doi.org/10.1007/978-3-642-42057-3_99
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-42056-6
Online ISBN: 978-3-642-42057-3
eBook Packages: Computer ScienceComputer Science (R0)